© Mary Ann Liebert, Inc. DOI: 10.1089/jwh.2006.0198
Avoidance of Environmental Tobacco Smoke among
Pregnant Taiwanese Women: Knowledge,
Self-Efficacy, and Behavior
CHING-MIN CHEN, R.N., D.N.S.,1 PI-HSIA LEE, R.N., Ed.D.,1 YU-HUA CHOU, R.N., M.S.N.,1 SHU-FEN KUO, R.N., M.S.N.,1
and YU-HIEN HSU, R.N., M.S.N.2
ABSTRACT
Background: The issue of active and passive cigarette smoking among pregnant women at home has become a major source of debate. The purpose of this study was to explore the knowledge of, self-efficacy with, and behavior toward avoiding environmental tobacco smoke and related factors among pregnant women in Taiwan.
Methods: A cross-sectional research design was used. Women (n 281) visiting the outpa-tient antenatal clinics of one regional hospital and two medical centers in Taipei for routine obstetrical care volunteered to fill out questionnaires.
Results: Participants on average had little knowledge of this issue and felt less than “very confident” in resisting environmental tobacco smoke and indicated that it was only “usually true” that they practiced avoidance behaviors. The knowledge of, self-efficacy with, and be-havior toward avoiding environmental tobacco smoke were all related to both the woman and her partner’s educational levels. There were significant differences in mean knowledge, self-efficacy, and avoidance of environmental tobacco smoke scores among different household smoking groups. A multiple regression revealed that overall avoidance of environmental to-bacco smoke was positively associated with self-efficacy, with a no-smoking policy at home, and with both a woman and her partner’s educational levels.
Conclusions: The high prevalence of subjects suffering from active (6.05%) and passive smoking (58.72%) suggests that clinicians can target interventions designed to increase preg-nant women’s self-efficacy and to advise them to try to set up their own smoking policy at home.
869 INTRODUCTION
S
MOKING IS ASSOCIATED with an increased riskof developing and of dying from many can-cers, cardiovascular disease (CVD), chronic ob-structive pulmonary disease (COPD), and more,
as well as an increased risk of adverse reproduc-tive outcomes.1 Maternal cigarette smoking
during pregnancy is associated with increased risk of miscarriage, intrauterine growth retarda-tion, preterm delivery, reduced infant weight, and higher infant mortality.1,2Many studies have 1Taipei Medical University College of Nursing, Taipei, Taiwan, R.O.C.
2Catholic Tien Nursing College, Taipei, Taiwan, R.O.C.
This research was supported by grant BHP-92-Anti-Tobacco-2U03 from the Bureau of Health Promotion, Depart-ment of Health, Taiwan.
also suggested that the pregnant woman’s expo-sure to her partner’s smoking may contribute to reduced birth weight3–5; paternal smoking also
increases the risk of infant respiratory infections and sudden infant death syndrome (SIDS), irre-spective of the maternal smoking status.3Efforts
to reduce prenatal cigarette smoke exposure have largely focused on encouraging women to quit smoking during pregnancy,6,7 but interventions
to promote maternal smoking cessation have not been very successful.4,8–10 Although women are
more likely to attempt to quit smoking during pregnancy than at other times in their lives, only one third stop successfully,9 and 50%–80% of
pregnant smokers relapse in the postpartum pe-riod.11 Research has found that family support
can be an important factor in contributing to pre-natal or early postpre-natal smoking cessation; fam-ily members’ own tobacco use, as well as their attitudes and behavior toward smoking by a pregnant women, can have a powerful influence on women’s smoking.5Exposure to daily passive
smoke at home (usually from a partner) is an im-portant risk factor for continued smoking during pregnancy, and support from the partner is the most important contributor to success in quit-ting.3 Targeting public relations efforts toward
the development of a smoke-free family has, therefore, become an important objective in many nations.7,12
Smoking is the leading source of indoor air pol-lution.13 The amount of environmental tobacco
smoke exposure, as measured by serum cotinine concentration in pregnant women, was statisti-cally negatively associated with the fetal bipari-etal diameter and the newborn child’s birth weight.14Thus, prenatal smoking cessation may
not eliminate health risks if women are exposed to smoke-laden environments. Sources of envi-ronmental tobacco smoke may include partners, friends, colleagues, and relatives who smoke and permissive rules about smoking in the home, car, and work environment. Although indoor tobacco smoking has been banned or significantly limited in many workplaces and public settings, the home is one of the only places with no external limits on smoking. Effective health promotion messages and programs must help develop a pregnant woman’s confidence in her ability to avoid environmental tobacco smoke from those living with her.
Women continue to smoke.15 In Taiwan,
ap-proximately 4.6% of women smoke during
preg-nancy despite the numerous adverse health con-sequences of smoking.16,17 Although the
preva-lence of maternal smoking in Taiwan may not be as high as it is in other countries,6more than half
of the pregnant women may be exposed to envi-ronmental tobacco smoke in their homes.16,18The
prevalence of cigarette use has been reported to be 50% for men aged 15 years and 58.3% for men aged 25–44 years; it is likely that the female partners of these men are of similar, reproductive age.18Recent studies support home smoking
re-strictions as being positively associated with in-creased attempts to quit, the intention to quit, lighter smoking, and sustained cessation.19–21
Further, smoke-free rule in homes can also reduce the risk of children becoming smokers,12yet
lit-tle is known about the extent to which pregnant women have control over avoiding environmen-tal tobacco smoke with the presence of other smokers in the home. The purpose of this study was to explore pregnant women’s self-efficacy in avoidance of environmental tobacco smoke. Re-sults of this study can inform pregnant women about the establishment of a smoke-free policy in the home to protect themselves and their unborn children. Although much has been published about the relationship between abstinence self-ef-ficacy and smoking cessation, only a few pub-lished reports have examined self-efficacy with regard to avoidance of environmental tobacco smoke.21
MATERIALS AND METHODS
Design
This was a cross-sectional study. All women who had a routine obstetrical appointment in the outpatient antenatal clinics of two medical cen-ters and one regional hospital were approached by one of three research assistants to take part in the study. Each woman was asked to complete a brief questionnaire in the waiting room prior to her appointment. The self-completion question-naire took 20–30 minutes to complete. Primary variables of interest included behaviors and self-efficacy in avoiding environmental tobacco smoke and knowledge about health risks and policies on smoking and smoking cessation. Both the ethics and methodology of the study proto-col were approved by the hospitals in question; all the women who volunteered for the study
pro-vided written consent. Results of a power analy-sis based on pilot test results indicated that a sam-ple of 243 women would have 80% power to de-tect an effect size of 0.2 (alpha 0.05).22
Subjects
Participants were recruited from three hospitals in Taipei. Flyers and information from clinic nurses were used to recruit women who received prenatal services between June and September 2004. Women were eligible for the study if they could read and write Chinese, were at least 16 years old, either 16, 28, or 38 weeks pregnant as per the recommended prenatal checkup times in the first, second, and third trimesters that would be covered by the National Health Insurance (NHI) program. The reason for selecting only wo-men at the noted weeks of pregnancy was for fu-ture planning for intervention in each gestational stage. Under the NHI program, only certain pre-natal checkups were covered by the plan. A gro-cery bag with wheels equal to a monetary reward of US$10 was offered as an incentive to participate.
Study variables/questionnaire
The questionnaires covered demographic char-acteristics, household smoking status, knowledge about health risks and policies on smoking and smoking cessation, and a self-reported measure of self-efficacy and avoidance of environmental tobacco smoke. Based on the work of Carmines and Zeller,23 development of this research
in-strument included focus group discussion to gen-erate items; content validity was established us-ing a panel of eight experts in nursus-ing, health education, and smoking cessation. The index of content validity (CVI) was determined by the proportion of experts who rated items as content valid (a rating of 3 or 4) using a 4-point ordinal rating scale. Finally, a pilot test of 30 pregnant women determined its face validity. Items were deleted based on correlations, variances, item-scale correlations, and alpha coefficients.24
The pregnant women’s avoidance of passive smoking was assessed using the Avoidance of En-vironmental Tobacco Smoke Scale.25The original
10-item scale was developed to characterize sub-jects’ behaviors associated with resisting envi-ronmental tobacco smoke and identifies situa-tions in which the exposure to environmental tobacco smoke occurs, such as, “If I encounter a friend who is smoking, I will still sit and talk with
him/her while he/she is smoking”, “When I am in a restaurant, I will leave if unable to sit in the nonsmoking section,” and “If I am with people who are smoking and I cannot leave, I will ask them to refrain from smoking.” This scale has been tested on college students and young moth-ers, and a known group comparison indicated a very high reliability and validity.25 The
ques-tionnaire used a 4-point scale for respondents to indicate their level of avoidance behavior (from 4, almost always true, to 1, almost never true). The scale was scored by averaging the answers, with the scores reversed for items worded nega-tively. In this study, two items were added espe-cially for pregnant women’s avoidance behaviors developed from focus group discussion: “When I encounter someone smoking outdoors, I will move away to avoiding exposure of smoke,” and “I don’t feel it is a problem to be in a smoking en-vironment.” After translation and back-transla-tion of the scale, the CVI was determined (CVI 0.98), and the overall reliability coefficient of 0.83 was comparable to the original published results of Martinelli.25
Social cognitive theory suggests that changes in behavior are influenced by a person’s self-effi-cacy or confidence in taking action in specific sit-uations.26The 6-item Avoidance of
Environmen-tal Tobacco Smoke self-efficacy scale was self-developed to assess how confident the wo-man was that she could keep others from smok-ing around her in public places, in others’ homes, and also in her own home. Response options for these items ranged from 0 (not at all confident) to 4 (extremely confident), producing mean scores of between 0 and 4, with higher scores rep-resenting higher confidence in avoiding environ-mental tobacco smoke. This scale was established to have an internal consistency coefficient of 0.83 and a CVI of 0.96.
The Knowledge of Smoking questionnaire contains 11 multiple-choice questions covering health risks of active and passive smoking on ma-ternal and fetal health as well as current anti-smoking regulations in Taiwan; three items re-quired multiple answers. The potential score ranged from 0 to 30 for this scale, and the score was converted to an accuracy proportion [(number of correct answers/30) 100], with a higher score in-dicating better knowledge. The validity and relia-bility of this scale (CVI 0.86, Kuder-Richardson (KR)-20 0.67) were also acceptable, as specified by DeVellis.24
Social demographic predictors of home smok-ing control included in the analyses were mater-nal age and both the woman and her partner’s educational levels and employment status. Infor-mation on the woman’s gestation and gravidity was also collected.
Analysis of data
In order to explore pregnant women’s knowl-edge of, self-efficacy with, and behavior toward avoiding environmental tobacco smoke, related factors, and predictors of avoidance behavior, data were collected through a survey method. The data were analyzed using the SPSS PC (ver-sion 10.0) statistical software package (SPSS, Chicago, IL). Descriptive statistics (frequency dis-tributions, means, and standard deviations [SD]) were used to characterize the study population. Unless otherwise specified, differences in the background characteristics and major outcome variables between smoking groups were exam-ined using chi-square test, Student’s t test, and analysis of variance (ANOVA) at the univariate level, and multiple linear regressions were used for the multivariate analysis.
RESULTS
Overall, 802 pregnant women were ap-proached; however, 282 did not meet the inclu-sion criteria because of language barrier or un-qualified weeks of pregnancy. Among 520 qualified subjects, 132 (25.38%) refused, and 107 (20.57%), although consenting, were unable to complete at least 50% of the questionnaire before they were called in for their appointment. This left a response rate of 54.04% (n 281). By chart review to compare any difference between non-participants and survey subjects, there was no significant difference in age and marital status.
Sample demographics and household smoking prevalence and patterns
The characteristics of the women in the survey are summarized in Table 1. The demographic characteristics of the study sample were similar to data retrieved from 8102 women who gave birth from 2002 to 2003.17Participants were
clas-sified into three household smoking status cate-gories based on response to a series of items
con-cerning the smoking status of members living in the same household. Ninety-four women (33.5%) were classified as living in the smoke-free fami-lies, indicating that they were nonsmokers and none of their family members in the same house-hold were current smokers. Among 187 pregnant women who lived in smoking households, 17 were active maternal smokers (refers to pregnant women who either did not abstain at all or had smoked during the week before the survey). The other 165 were passive smokers who were non-smokers but lived with non-smokers in their house-hold who may have been the spouse/partner (n 138), other family members living in the same household (n 12), or both the partner and other families living together (n 15). Among women with partners who smoked (n 153), the majority (56.9%) reported that they had cut down on the number of cigarettes smoked since the wo-man’s pregnancy, as had 50.5% of the other fam-ily members who smoked. Regarding household smoking regulations, subjects were asked to se-lect one answer to best describe their household smoking situation: smoking was not allowed in any place, smoking was allowed in certain des-ignated areas, or there is no smoking regulation at home. Only 43.7% of the subjects stated that there was a strict no-smoking policy in their home (Table 2).
TABLE1. CHARACTERISTICS OF SAMPLE(N 281)
Characteristic Frequency (%) Education (n 279) High school 17 (6.1) High/vocational school 89 (31.9) College 81 (29.0) University or higher 92 (33.0) Spouse/partner’s education (n 277) High school 17 (6.1) High/vocational school 79 (28.5) College 68 (24.5) University 75 (27.1) Postgraduate 38 (13.7) Trimester First 65 (23.1) Second 107 (38.1) Third 109 (38.8) Gravidity Primigravida 154 (54.8) Multigravida 127 (45.2) Working situation (n 234) Employed 130 (55.6) Unemployed 104 (44.4) Age (mean SD) 29.66 4.83
There was no significant variation in subjects’ characteristics by household smoking status ex-cept for both women and their partners’ educa-tional levels. A higher educaeduca-tional background in both women (chi-square (5, n 279) 30.31, p 0.000) and their partners (chi-square (5, n 277) 38.66, p 0.000) tended to represent higher proportions of smoke-free families. Cross-tabulation indicated that 71.8% of pregnant wo-men in the low-education group did not have environmental tobacco smoke control set up in their homes (chi-square (5, n 275) 16.05, p 0.000), and similar results were found for these women’s spouses/partners’ educational back-ground (chi-square (5, n 273) 31.86, p 0.000).
Factors related to knowledge of, self-efficacy with, and behavior toward avoiding environmental tobacco smoke
Pregnant women’s knowledge of smoking was assessed to evaluate subjects’ understanding or denial of health risks and policies. On average, women only scored 53.26% (SD 14.99), indicating that most women did not have a strong under-standing of the health risks of active and passive
smoking on maternal and fetal health or of cur-rent antismoking regulations in Taiwan. Among the adverse health consequences of smoking, 80% of subjects did not know of the increased risks of developing breast cancer and premature menopause for women and higher odds of de-veloping dental cavities and infantile colic in their children.
Average scores on the 6-item avoidance of en-vironmental tobacco smoke self-efficacy ranged from 0 to 4 (mean SD, 2.97 0.85), indicating that pregnant women felt less than “very confi-dent” in resisting environmental tobacco smoke. In terms of pregnant women’s behaviors in avoid-ing environmental tobacco smoke, the average score of 3.09 (SD 0.52, range 1.42–4) indicated that women believed it to be “usually true” that they avoid environmental tobacco smoke. Women who were nonsmokers practiced better avoidance behavior (mean SD, 3.37 0.41) than passive smokers (mean SD, 3.00 0.48) and smokers (mean SD, 2.39 0.51).
Knowledge of, attitudes toward, and behavior toward avoidance of environmental tobacco smoke were significantly related to both the wo-men and their partners’ educational levels. Preg-nant women with higher educational preparation
TABLE2. PATTERNS ANDSOURCES OFACTIVE ANDPASSIVESMOKE OFSAMPLE
Variable Frequency (%)
Household smoking status (n 281)
Smoke-free family 94 (33.5)
Smoking family 187 (66.5)
Source of smoking (n 187)
Maternal active smoke 12 (6.7)
Both active and passive 5 (2.7)
Passive smoke 165 (90.6)
Missing 5 (N/A)
Women’s smoking status (n 280)
Never smoked 231 (82.5)
Quit before pregnancy 4 (1.4)
Quit after pregnancy 28 (10)
Current smoker 17 (6.1)
Spouse/partner who smoked (n 153)
Smoked as usual 64 (41.8)
Smoked less after pregnancy 87 (56.9)
Smoked more after pregnancy 2 (1.3)
Other householder’s smoking status (n 97)
Smoked as usual 48 (49.5)
Smoked less after pregnancy 49 (50.5)
Household smoking regulation (n 277)
Not allowed 121 (43.7)
In designated areas 104 (37.5)
No regulation 52 (18.8)
not only knew more about smoking risks and an-tismoking laws (rs 0.35, p 0.01) and were more confident about their ability to control pas-sive smoke (rs 0.18, p 0.01) but also took more concrete actions to avoid environmental to-bacco smoke (rs 0.43, p 0.01) than those with lower educational preparation. A consistent pat-tern was observed in knowledge of (rs 0.29, p 0.01), self-efficacy with (rs 0.23, p 0.01), and behavior toward (rs 0.48, p 0.01) the wo-men’s partners. Maternal age was positively as-sociated with avoidance behavior (r 0.16, p 0.01); employed women (mean SD, 3.13 0.47) also performed better in avoiding environmental tobacco smoke than those unemployed (mean SD, 2.99 0.59) (t 2.18, p 0.05). Knowledge about environmental tobacco smoke was higher among primigravida women (mean SD, 55.92 13.86) than multigravid ones (mean SD, 50.08 14.64) based on Student’s t test analy-sis (t 3.40, p 0.001). Stages of pregnancy (trimester) were not associated with either the knowledge of, self-efficacy attitude toward, or be-havior toward avoiding environmental tobacco smoke.
To explore the differences in knowledge of, self-efficacy attitude toward, and behavior to-ward avoiding environmental tobacco smoke among different household smoking groups, ANOVA was applied. Overall, mothers who con-tinued to smoke were less likely to be aware of, or convinced of, the dangers of smoking (F(2, 271) 7.11, p 0.01) and were less confident (F(2, 270) 5.91, p 0.01) and capable in
avoid-ing environmental tobacco smoke (F(2,273) 40.19, p 0.001) than women who suffered from passive smoking or those without household to-bacco smoke exposure. There were also signifi-cant differences in knowledge of (F(3, 270) 5.07,
p 0.01), self-efficacy with (F(3, 269) 4.02, p
0.01), and behavior toward (F(3, 272) 28.37, p 0.01) avoiding environmental tobacco smoke at the source of smoking. Women in smoke-free family groups performed the best, and active ma-ternal smokers had the lowest scores on both scales. However, when the source of smoking was from their husbands or partners, women had better self-efficacy but had worse avoidance be-havior than when the passive smoke was from other family members. Finally, women with a no-smoking home policy had better knowledge (t 2.39, p 0.05), self-efficacy (t 3.19, p 0.05), self-efficacy (t 3.19, p 0.01), and avoid-ance behaviors (t 7.60, p 0.000).
Factors predicting avoidance of environmental tobacco smoke
In order to identify the variables most strongly related to pregnant women’s practice of avoiding passive smoking, each significant variable in the bivariate analysis was entered into a multiple re-gression (Table 3). After adjusting for age, em-ployment, and education levels (both the women and their partners) (model 1), self-efficacy of re-sistance to passive smoke significantly predicted the pregnant women’s behavior in avoiding en-vironmental tobacco smoke (F 25.92, p
TABLE3. MULTIPLEREGRESSION OFMAJORVARIABLES ONPREGNANTWOMEN’S AVOIDANCE OF ENVIRONMENTALTOBACCO SMOKE(N 281)
Model 1 Model 2 Model 3
Predictor variable t t t Age 0.05 0.86 0.08 1.54 0.06 1.20 Employment 0.01 0.21 0.03 0.51 0.02 0.32 Woman’s education 0.32 4.34*** 0.25 3.65*** 0.24 3.70** Partner’s education 0.25 3.62*** 0.21 3.31** 0.15 2.32* Self-efficacy 0.40 7.52*** 0.38 7.30*** Knowledge 0.02 0.36 0.00 0.03
Home smoking policy 0.19 2.14*
Household smoking status 0.06 0.70
Constant 2.51 1.74 1.74
Adjusted R2 0.25 0.40 0.45
F 20.23 25.92 23.47
p 0.000 0.000 0.000
0.000). This variable explained 15% of the vari-ance; knowledge, however, was excluded from the regression model (model 2).
In regression model 3, the home smoking pol-icy and household smoking status were trans-formed into dichotomous variables and added to the regression. Overall avoidance of environ-mental tobacco smoke was significantly posi-tively associated with self-efficacy, with a no-smoking policy at home, and with both women and their partners’ educational levels (F 23.47,
p 0.000). A home smoking policy explained 5%
of the variance; household smoking status, how-ever, made no significant contribution to the mul-tiple regression model.
DISCUSSION
As it is possible to spontaneously stop smok-ing and there is high motivation for women to al-ter this behavior, the potential for a positive health impact is great during pregnancy, and many programs have targeted smoking cessation by pregnant women.27 Interventions and
provi-der advice that focus solely on maternal behav-ior do not address the full range of smoking risks, however, given that women are at continued risk of exposure to passive smoking. This may be problematic among women who quit smoking for their pregnancies but who live in an environment that is permissive of others’ smoking. Therefore, it is important to address the problem of envi-ronmental tobacco smoke exposure during preg-nancy. The present study examined factors asso-ciated with pregnant women’s behavior in avoiding environmental tobacco smoke. Factors related to pregnant women’s behavior in resist-ing active and passive smokresist-ing need to be fur-ther identified in order to develop better strate-gies for minimizing and avoiding exposure. Information on the success of interventions to re-duce environmental tobacco smoke exposure of pregnant women is limited.
In our study, the prevalence of women exposed to active (17 of 281 6.05%) and passive smok-ing (165 of 281 58.72%) was consistent with the results of Lin,17 who surveyed 12,857 women
who gave birth in seven municipal hospitals in Taipei. Lin17found that 44% of all pregnant
wo-men had been exposed to second-hand smoke from their spouses, and an additional 10% had been exposed from other family members. Lin’s
findings (54%) were slightly lower than ours (58% 72%), which may have been due to the nature of the convenience sampling we used and which may have overrepresented the high prevalence of passive smokers who enrolled in our study. Al-though pregnant women suffering from passive smoke practiced better behaviors in avoiding en-vironmental tobacco smoke than did smokers, as indicated in our results their risks were consis-tently higher than those of nonsmokers. An as-tonishing 54%–66.5% of the women resided with a smoker and were potentially exposed to the ad-verse health effects of second-hand smoke. These women were trapped in an environment that ex-posed them to multiple health problems not only in terms of the pregnancy and health outcomes of the infant but also in terms of coronary artery disease, stroke, and various types of cancer. How-ever, 80% of the sample we surveyed were not aware of the long-term health risks of active and passive smoking.
Other studies have also supported the need to address smoke exposure during pregnancy.8,11
Overall, smokers were less likely to be convinced of the dangers of smoking related to pregnancy and child development than were those exposed to passive smoking or who lived in a smoke-free household. This may imply a knowledge deficit on the part of continuing smokers, and it may also reflect a belief by such women that these risks are unlikely, possibly influenced by their observation that smokers frequently have normal birth weight babies.28It may be advisable to deemphasize the
immediate risks to the mother and fetus and stress other aspects of smoking that may be of more personal relevance and carry long-term health risks to women and children, such as can-cer and cardiovascular disease.
In the present study, 43.7% of the women claimed to have set up an antismoking policy in their homes, which is lower than the proportion of persons who were covered by smoke-free home rules, which ranged from 51% in Kentucky to 86% in Utah, in the United States.12Other
stud-ies have reported prevalences of home smoking restrictions ranging from 12.5% to 43% in homes with children and at least one smoker.21Dunn et
al.29 found that the percent of pregnant women
who allowed smoking in their homes was 31.7% of nonsmokers, 46.9% of abstainers, and 84.7% of smokers. A consistent pattern was observed in ex-posure to the partner’s second-hand smoke across smoking categories, and education levels
were associated with second-hand smoke expo-sure. In our study, both women and their part-ners’ educational levels were associated with the household smoking status and the presence of a smoking policy at home. Women with a higher educational background usually marry someone with an equivalent educational level, and these couples represented a higher proportion of non-smokers themselves; also, a higher proportion lived in nonsmoking households. Results of this study indicate that these women were more con-fident in setting up smoking restrictions in their households than were women with a high school or lower education.
According to social cognitive theory, self-effi-cacy is a predictor of behavior.26Consistent with
other studies,21confidence in one’s ability to
con-trol smoking in the living environment was a sig-nificant predictor of avoidance of environmental tobacco smoke in this study. Few other studies have explored exposure avoidance self-efficacy among parents of infants and children. Strecher et al.30also found that a mother’s confidence in
her ability to protect her children from environ-mental tobacco smoke varied as a function of the setting and the individuals with whom the mother must intervene to limit exposure. Crone et al.31reported that mothers found it more
dif-ficult to ask family or friends than to ask their partners not to smoke in the presence of the in-fant. In this study, pregnant women revealed bet-ter self-efficacy in avoiding environmental to-bacco smoke from their partners but higher tolerance of it from their husbands or partners than from other family members.
The multivariate analysis also revealed that en-vironmental tobacco smoke avoidance behaviors were dramatically higher among the women who have higher self-efficacy and had a no-smoking policy at home. Because of the nature of a cross-sectional study design, the present study does not allow conclusions with regard to cause and effect. However, pregnant women suffering from pas-sive smoking might strengthen their confidence in response to someone who smoked in their household by setting up a no-smoking policy. The findings of this study suggest that health pro-fessionals and smoking cessation programs should be more aggressive in targeting women’s self-efficacy and in instructing them on how to set up a smoking policy at home, as these two fac-tors positively predicted the actual performance of avoiding environmental tobacco smoke among
pregnant women. In line with this recommenda-tion, self-efficacy can play a major role in deter-mining the degree to which women can avoid passive smoke. The U.S. Surgeon General has concluded that eliminating smoking in indoor spaces is the only way to fully protect nonsmok-ers from second-hand smoke exposure.12 Thus,
interventions that target pregnant women alone would be problematic if the woman lives with a smoker. Findings of this study are particularly important for policymakers to consider strength-ening health education on the importance of this issue to women’s spouses and other family mem-bers. Through this widespread effort, a consen-sus can be formed to help set up total smoking bans in these households in order to create a sup-portive healthy environment rather than leaving pregnant women to fight on their own.
Limitations
This study used a convenience sample of wo-men who were either 16, 28, or 38 weeks preg-nant and sought routine obstetrical care at three urban hospitals, and the findings might not be able to be generalized to women at other preg-nant stages or in other settings. There is always the possibility when respondents complete a forced choice questionnaire that questions may be misinterpreted. Nonparticipants could not be characterized because of the passive nature of re-cruitment. The extent of the active and passive exposures was not assessed to verify the psycho-metric properties of the scales used in this study because of resource limitations.
CONCLUSIONS
Actions to limit smoking inside the home are being undertaken by many pregnant women to try to limit environmental tobacco smoke expo-sure in lieu of or in conjunction with smoking ces-sation. The results of the present exploratory study suggest that establishment of a policy against smoking in the home might produce the best chance for women to actually avoid envi-ronmental tobacco smoke. Pregnant women’s at-titudes about their ability to control exposure appear to be predictors of active environmental tobacco smoke control. Low self-efficacy might be amenable to intervention. Clinicians are encour-aged to screen for environmental tobacco smoke
exposure from household members and other in-dividuals in the pregnant women’s environment. Clinicians should explore with pregnant women how confident they feel in their ability to enforce a policy that limits or eliminates smoking around them. More research is needed to determine ef-fective strategies for facilitating efforts made by families to ensure a smoke-free environment.
Prenatal smoking cessation will not eliminate health risks if women continue to be exposed to second-hand smoke. This study compared the knowledge of, self-efficacy with, and behavior toward avoiding environmental tobacco smoke among pregnant women in Taiwan. The find-ings provide further empirical support for the implication that women’s self-efficacy and avoidance of environmental tobacco smoke are correlated with family smoking status, and dif-ferent interventions are needed to meet preg-nant women’s needs. It is important that addi-tional risks of environmental tobacco smoke exposure be addressed through a more compre-hensive approach to prenatal smoking, such as biochemical verification of the level of exposure. Clinicians can also determine the prevalence of smoking and environmental tobacco smoke ex-posure among their clientele by implementing a survey. Hence, early interventions that begin raising awareness about the risks of environ-mental tobacco smoke exposure during preg-nancy and that help women develop skills for negotiating situations where smoking occurs would benefit maternal and prenatal health. An-tismoking counseling should, therefore, be in-corporated into prenatal counseling and given as a standard procedure at the beginning of ob-stetrical care.
ACKNOWLEDGMENTS
We thank Dr. Jane Robinson for her careful re-view of the manuscript, and we also extend thanks to the women who participated in the sur-vey.
REFERENCES
1. Ventura SJ, Hamilton BE, Mathews TJ, Chandra A. Trends and variations in smoking during pregnancy and low birth weight: Evidence from the birth cer-tificate, 1990–2000. Pediatrics 2003;111:1176.
2. Salihu HM, Aliyu MH, Pierre-Louis BJ, Alexander GR. Levels of excess infant deaths attributable to ma-ternal smoking during pregnancy in the United States. Matern Child Health J 2003;7:219.
3. Wakefield M, Beid Y, Roberts L, Mullins R, Gillies P. Smoking and smoking cessation among men whose partners are pregnant: A qualitative study. Soc Sci Med 1998;47:657.
4. Pletsch P, Morgan S. Smoke-free families: A tobacco control program for pregnant women and their fam-ilies. Journal of Obstetric Gynecologic and Neonatal Nursing (JOGNN) 2002;31:39.
5. McBride CM, Curry S, Grothaus LC, Nelson JC, Lando H, Pirie PL. Partner smoking status and preg-nant smoker’s perceptions of support for and likeli-hood of smoking cessation. Health Psychol 1998; 17:63.
6. Albrecht SA. Achieving “success”: Nursing care for pregnant women who smoke. Association of Women’s Health, Obstetric, and Neonatal Nurses (AWHONN) Lifelines 2004;8:190.
7. Bureau of Health Promotion, Department of Health. BHP to create “smoke free” Taiwan [monograph on the Internet]. Taipei: Bureau of Health Promotion, 2004. Available at www.doh.gov.tw/dohenglish/media/ se20040611-1.asp Accessed April 10, 2005.
8. Windsor RA, Boyd NR. A meta-evaluation of smok-ing cessation intervention research among pregnant women: Improving the science and art. Health Educ Q 1998;13:419.
9. Klesges LM, Johnson KC, Ward KD, Barnard M. Smoking cessation in pregnant women. Obstet Gy-necol Clin 2001;28:269.
10. Lancaster T, Stead L, Silagy C, Sowden A. Effective-ness of interventions to help people stop smoking: Findings from the Cochrane library. BMJ 2000;321: 355.
11. Oncken CA, Kranzler HR. Pharmacotherapies to en-hance smoking cessation during pregnancy. Drug Al-cohol Rev 2003;22:191.
12. United States Department of Health and Human Ser-vices. Press release: New Surgeon General’s report fo-cuses on the effects of secondhand smoke. Atlanta, GA: U.S. Department of Health and Human Services, 2006. Available at www.hhs.gov/news/press/2006pres/ 20060627.html Accessed April 10, 2005.
13. National Cancer Institute. Health effects of exposure to environmental tobacco smoke: The report of the California Environmental Protectional Agency. Bethesda, MD: National Cancer Institute, 1999. 14. Hanke W, Sobala W, Kalinka J. Environmental
to-bacco smoke exposure among pregnant women: Im-pact on fetal biometry at 20–24 weeks of gestation and newborn child’s birth weight. Int Arch Occup Envi-ron Health 2004;77:47.
15. Sheahan SL, Free TA, Rayens MK. Smoking behavior and desire to quit among low-income female care-givers. Appl Nurs Res 2003;16:156.
16. Chen GW. Smoking behaviors and knowledge to-ward health risks of it among pregnant women in
Tai-wan. Final report. Taipei, Taiwan: Department of Health, 1993. Report No. DOH82-HP-088-3M24. Con-tract No. DOH82-HP-088-3M24.
17. Lin CL. The effect on perinatal morbidity and mor-tality for pregnant women or their spouse with smok-ing. Final report. Taipei, Taiwan: Bureau of Health Promotion, 2004. Report No. BPH-92-Anti-Tobacco-F202. Contract No. BPH-92-Anti-Tobacco-BPH-92-Anti-Tobacco-F202. 18. Bureau of Health Promotion [database on the
Inter-net]. Population and health survey. Research Infor-mation Sharing System. c2005. Available at rds. bhp.doh.gov.tw/www/index.jsp Accessed April 1, 2005. 19. Irvine L, Crombie IK, Clark RA, et al. Advising par-ents of asthmatic children on passive smoking: Ran-domised controlled trial. BMJ 1999;318:1456.
20. Okah FA, Choi WS, Okuyemi KS, Ahluwalia JS. Ef-fect of children on home smoking restriction by in-ner-city smokers. Pediatrics 2002;109:244.
21. Sockrider MM, Hudmon KS, Addy R, Mullen PD. An exploratory study of control of smoking in the home to reduce infant exposure to environmental tobacco smoke. Nicotine Tobacco Res 2003;5:901.
22. Cohen J. Statistical power analysis for the behavioral science. Mahwah, NJ: Erlbaum Association, 1988 23. Carmines FC, Zellar RA. Reliability and validity
as-sessment. Beverly Hills, CA: Sage, 1979
24. DeVellis R. Scale development: Theory and applica-tions. Newbury Park, CA: Sage, 1991.
25. Martinelli AM. Development and validation of the Avoidance of Environmental Tobacco Smoke Scale. J Nurs Meas 1998;6:75.
26. Bandura A. Social foundation of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Pren-tice-Hall, 1986.
27. Office of the Surgeon General. Tobacco cessation guideline. Rockville, MD: U.S. Public Health Service Publication, 2000.
28. Wakefield MA, Psych DA, Jones WR. Cognitive and social influences on smoking behaviour during preg-nancy. Aust NZ J Obstet Gynaecol 1991;31:235. 29. Dunn CL, Pirie PL, Hellerstedt WL. Self-exposure to
secondhand smoke during prenatal smokers, abstain-ers, and nonsmokers. Am J Health Prom 2004;18:296. 30. Strecher VJ, Bauman KE, Boat B, Fowler MG, Green-berg RA, Stedman H. The development and forma-tive evaluation of a home-based intervention to re-duce passive smoking by infants. Health Educ Res 1989;4:225.
31. Crone MR, Rejneveid SA, Burgmeijer RJF, Hirasing RA. Factors that influence passive smoking in infancy: A study among mothers of newborn babies in The Netherlands. Prev Med 2001;32:209.
Address reprint requests to:
Ching-Min Chen, R.N., D.N.S. Taipei Medical University College of Nursing 250 Wu-Hsing Street Taipei, Taiwan, R.O.C. E-mail: [email protected]
1. S. Q. Lu, R. Fielding, A. J. Hedley, L.-C. Wong, H. K. Lai, C. M. Wong, J. L. Repace, S. M. McGhee. 2011. Secondhand Smoke (SHS) Exposures: Workplace Exposures, Related Perceptions of SHS Risk, and Reactions to Smoking in Catering Workers in Smoking and Nonsmoking Premises. Nicotine & Tobacco Research . [CrossRef]
